U N D E R G R A D U A T E R E S E A R C H O P P O R T U N I T Y P R O G R A M
Cognitive and Experimental Economics
Undergraduate Research Opportunity Program (UROP) – Ann Arbor, Mich.
Undergraduate Researcher, September 2008 – May 2009
Social science research in “Cognitive and Experimental Economics.” Studied decision-making over time, the extent of naïvete in time discounting and its relation to cognitive ability. Read several research papers on the subject. Improved existing survey, distributed surveys to small test group over time, collected and inputted data for analysis. Began preparations to conduct a larger experiment.
S Y N O P S I S
In our study, we ask what the extent of naïvete in time-discounting is, and link it to cognitive ability. Large bodies of research are based on the assumption that people discount future benefits and costs. Most economists modeling time-inconsistent preferences assume sophistication - not naïvete. Sophistication implies that people have “rational expectations” about future behavior.
Exponential discounters, all of whom are sophisticated, value preferences in a time consistent manner.
Quasi-hyperbolic discounters are present-biased, delaying immediate-cost activities and engaging in immediate-reward activities (O’Donoghue and Rabin “Doing It Now or Later” The American Economic Review, Vol. 89, No. 1, pp. 103-124).
Because they suffer from time-inconsistency, quasi-hyperbolic discounters can be naïve or sophisticated.
There is limited research on naïvete with weak anecdotal evidence, such as procrastination. For instance, people always will plan to quit smoking tomorrow - but never do.
Thus far, there has been no testing of the extent of naïvete in a unified framework. In order to identify the correlation between naïvete and different types of time-discounting, participating college students took the same survey twice, separated by two weeks. The survey consists of four sections. The first three offered students a payment of $10 and payment options >$10 at later times. The fourth asks for demographic information, GPA and standardized test scores. GPA and standardized test scores were used to assess relative cognitive ability.
Survey uses real stakes (i.e. cash) that are paid to respondents based on their answers to the questions. This is to ensure respondents’ answers are similar to actual behavior.
We expect that a significant portion of our sample will be naïve in their discounting. We also expect a positive correlation between mathematical intelligence and sophistication. Specifically, we hypothesize:
By conducting this study, we hope to provide unequivocal evidence of the correlation between naïvete and time-discounting type. This will have important implications for research and public policy which enforces commitment in areas such as addiction, savings and retirement. If people are naïve, polices might be better aimed at either making them more sophisticated, or providing incentives for people to use commitment devices.
Undergraduate Researcher, September 2008 – May 2009
Social science research in “Cognitive and Experimental Economics.” Studied decision-making over time, the extent of naïvete in time discounting and its relation to cognitive ability. Read several research papers on the subject. Improved existing survey, distributed surveys to small test group over time, collected and inputted data for analysis. Began preparations to conduct a larger experiment.
- Undergraduate Research Opportunity Program Symposium (May 2009, Ann Arbor, MI)
S Y N O P S I S
In our study, we ask what the extent of naïvete in time-discounting is, and link it to cognitive ability. Large bodies of research are based on the assumption that people discount future benefits and costs. Most economists modeling time-inconsistent preferences assume sophistication - not naïvete. Sophistication implies that people have “rational expectations” about future behavior.
Exponential discounters, all of whom are sophisticated, value preferences in a time consistent manner.
- If tomorrow one apple is worth 1/(1+i) apples today, then a year from now, tomorrow one apple is worth 1/(1+i) apples today. (E.g. One apple today vs. Two apples tomorrow & One apple in one year vs. Two apples in one year plus one day)
Quasi-hyperbolic discounters are present-biased, delaying immediate-cost activities and engaging in immediate-reward activities (O’Donoghue and Rabin “Doing It Now or Later” The American Economic Review, Vol. 89, No. 1, pp. 103-124).
- If one apple today is selected now, and when exactly the same choice is given 365 days later, two apples in one year plus one day is selected. (E.g. One apple today vs. Two apples tomorrow & One apple in one year vs. Two apples in one year plus one day)
Because they suffer from time-inconsistency, quasi-hyperbolic discounters can be naïve or sophisticated.
- Naïve people are influenced solely by a present-biased effect, and assume they will act in a time consistent manner in the future. They believe that they will behave themselves in the future.
- Sophisticated people act with full knowledge of their time inconsistency. They know they may not behave themselves due to a self-control problem.
There is limited research on naïvete with weak anecdotal evidence, such as procrastination. For instance, people always will plan to quit smoking tomorrow - but never do.
Thus far, there has been no testing of the extent of naïvete in a unified framework. In order to identify the correlation between naïvete and different types of time-discounting, participating college students took the same survey twice, separated by two weeks. The survey consists of four sections. The first three offered students a payment of $10 and payment options >$10 at later times. The fourth asks for demographic information, GPA and standardized test scores. GPA and standardized test scores were used to assess relative cognitive ability.
- A=Today, B=One Month
- A=One Month B=Two Months
- A=One Month, B=Two Months
- Demographic information, GPA, standardized test scores
Survey uses real stakes (i.e. cash) that are paid to respondents based on their answers to the questions. This is to ensure respondents’ answers are similar to actual behavior.
We expect that a significant portion of our sample will be naïve in their discounting. We also expect a positive correlation between mathematical intelligence and sophistication. Specifically, we hypothesize:
- 70-90% of individuals will exhibit quasi-hyperbolic discounting (50-80% naïve and 10-20% sophisticated)
- 10-30% of individuals will exhibit exponential discounting
By conducting this study, we hope to provide unequivocal evidence of the correlation between naïvete and time-discounting type. This will have important implications for research and public policy which enforces commitment in areas such as addiction, savings and retirement. If people are naïve, polices might be better aimed at either making them more sophisticated, or providing incentives for people to use commitment devices.
